Keywords: Remote Sensing, Land Use, PCA, TSA, NDVI

Abstract
This study reveals some of temporal-spatial dynamics of land-cover change in China from 1982 to 1999 through Principal Components Analysis (PCA) of Pathfinder AVHRR Land (PAL) normalized difference vegetation index (NDVI) data. A more overall and newer understanding of China's land-cover change dynamic can be achieved by a longer time series analysis (TSA). The results have proved that PCA/TSA is a very effective method to identify both macro and micro factors driving the change of NDVI. Especially the thesis to paves a way to detect the impacts of extreme physical accidents and human-induced activities upon the NDVI change. The analytical results are quite exciting and satisfactory.

Introduction
China is the world's third largest country by land area, the largest in population, and by some accounts, the third largest economy. In the past two decades, China's economy has experienced radical change, with double-digit growth rates in some years. As a result, the land use/cover change is also extensive. Rapid urbanization is leading to the loss of agricultural land. According to China's land resources management agency, the cultivatable acreage decreased by approximately 0.9% per year during the 1980s and early 1990s (Chen and Qiu, 1994). Moreover, desertification results in the gradual encroachment of originally vegetated areas. Meanwhile, some large catastrophic events, such as great floods and droughts, as well as increasing environmental pollution, may well have devastated vulnerable ecosystems, bringing about irreversible alteration of the vegetative cover. On the other hand, reforestation is, to some extent, potentially beginning to offset the negative impact of deforestation. It is apparent that the dynamic of land-cover change in China is quite complicated, because it is driven by biophysical, political, economic, and even cultural factors. China is, and will be, responsible for significant environmental changes at local to global scales (He, 1991; Young and Wang, 2000). Therefore, it is necessary to study China's land-cover change dynamic, and to find appropriate strategies to reduce the negative impacts of environmental changes.

This study reveals some of temporal-spatial dynamics of land-cover change in China from 1982 to 1999 through Principal Components Analysis (PCA) of Pathfinder AVHRR Land (PAL) normalized difference vegetation index (NDVI) data. It has been proved that unispectral time series remote sensing data can be used effectively to identify extensive and subtle changes of NDVI by means of PCA (Young, 1997). PCA is also suitable for removing noise within a multispectral image (Young, 1997). Analysis of unispectral PCA for China's vegetation change was performed by Young (1998), who averaged monthly composites to create eleven annual mean NDVI images for 1982 through 1992. He found some significant land-cover change patterns. Yet, a further study deserves undertaking, as a larger dataset with approximately 20-year coverage replaces the old dataset. A broader, updated understanding of China's land-cover change dynamic can be achieved by a longer time series analysis. Doing so, the following objectives can be expected. First, an overall spatial pattern and temporal trend of NDVI change for all of China can be gained by the yearly average NDVI value analysis and its standard variation analysis over the past two decades. Further, the difference of NDVI temporal-spatial patterns of different phenological regions can be examined. Second, using the yearly average NDVI value analysis within specific sub-regions, such as the Yangtze River Valley, the Yellow River Valley, the Pearl River Delta and Northeast China, some more subtle but sensitive changes in vegetative cover can be ascertained, such as the impact of natural disasters as well as man-made changes.

Metrology

Data
This study uses the National Oceanic and Atmospheric Administration (NOAA) / National Aeronautics and Space Administration (NASA)'s Pathfinder AVHRR Land data set with 8 km resolution (Agbu and James, 1994).

Analytical Technique
Principal Components Analysis (PCA) is an orthogonal transformation of n-dimensional image data that produces a new set of images. Known as components, these images are uncorrelated with one another and are ordered in terms of the amount of variance they explain in the original data set.

Framework
The purpose of this study, as addressed above, is intended to detect the temporal-spatial pattern of NDVI change in China and the possible natural and human agents responsible for these changes. Therefore, the author has designed a study framework following two main threads; that is, from macro to micro, as well as from the general to the specific, to detect these variations.

First, an overall analysis will be performed for all of China, the purpose of which is to investigate the general temporal trends and spatial distributions of NDVI within the entire country, and further to detect if there are possible regional differences in different sub-regions by means of the principal components analysis of the imagery.

Second, if regional differences in NDVI change patterns can be found, some further analyses for characteristic sub-regions will be undertaken. These sub-regions will be chosen for their physical phenological conditions, such as climate and hydrological conditions, as well as for their significance to China with respect to their agricultural importance and population characteristics.

Finally, more detailed analyses will be undertaken to further identify the impacts of natural catastrophes and human interventions upon the change of vegetation land-cover within the sub-regions through specific PCA. The study will especially focus on flood hazards and urbanization phenomena.

Study Areas

China Nation-Wide
As the third largest country in the world, China is so large that its territory spans multiple climatic zones, from the tropical and subtropical, to the temperate, to frigid zones, to the specific high-frigid climate style on high elevation plateaus. Taken as a whole, the vegetation types in China are very rich and versatile, with a nation-wide distribution geographically. In addition, with extensive human development over millennia, land use and land cover have experienced profound changes.

The study considers the entire area of China as well as its three sub-regions as the study areas, that is, Northeast China, the Yellow River Valley, and the Yangtze River Valley.

Northeast China
The northeastern area is the furthest northern part as well as the largest natural forest zone in China, where the climate is frigid and semi-humid. Natural vegetation there has been well maintained due to relatively less human development compared to the southern part of China. Biophysically, this area is an independent geographic division in terms of its climate and hydrological conditions.

Yellow River Valley
The Yellow River is the cradle of Chinese civilization. Human development, as well as the conversion of natural conditions over the past five millennia, has deeply transformed the landscape of this area. As a result, the natural vegetation coverage is much less than in ancient times, and much of the land has been converted to agricultural use. Monthly PCA/TSA was applied to this area for detecting the present-day pattern of NDVI change.

Yangtze River Valley
The Yangtze River is the longest river in China, beginning in Qinghai province located on the world's highest plateau, and crossing over almost the entire continent easterly to the sea. The Yangtze River Valley is the most vigorous and important economic zone of China. Here, the LiangHu plain of the Hubei and Hunan provinces is well known as the famous "Cereal Depository", and the city of Shanghai at the river's estuary is China's economic and financial center, playing the leading role in a new round of Chinese economic reform and development. However, the fertile Yangtze River Valley also suffers frequent natural hazards. Further, overdevelopment (its population is about 40% of the entire nation (Huang, 1994)) and unsustainable practices such as agricultural expansion by means of constructing closed levees directly cause increased soil erosion and sediment accumulation in river courses, aggravating the recurrence and magnitude of hazards (Cai, 1992). The following section of the thesis will analyze and explain the impacts of natural disasters and human-induced activities upon regional vegetation cover change through NDVI analysis.

Lake Dongting
Lake Dongting is famous for its cereal and fish production in China. Although its cropland area is just over 2% of the nation, its cereal production is 6% of the nation's total output (Huang, 1994). However, this area is also a hazard zone with very frequent floods as well as droughts. In the past 50 years this area has been subjected to 38 flood disasters, which brought huge economic losses and deaths (Li and Duan, 1997). Besides regional physical and hydrological conditions that cause flood hazards, (e.g., its lower elevations and its four big branches at the confluence of the Yangtze River), human factors have increasingly been contributing to the formation of the disasters. The most significantly negative human development is the so-called "create croplands by encroaching the lake area" phenomenon (Xiang, 1999). Due to the creation of new croplands, the lake area has dramatically shrunk 1, 659 km2 (Wang, 1999), in turn further triggering large amounts of sediment accumulation and river course blockage, and aggravating the threat of flood disasters.

For natural hazard analysis, I concentrate on the Lake Dongting region. In the 1990s, except 1990, 1992, and 1997, floods occurred every other year. In particular, the floods of 1996 and 1998 created new historical records (Xiang, 1999). For this, the author selected the NDVI data of this area ranging from 1995 to 1999, and attempted to identify the flood accidents as well as the possible submerged extent using PCA.

Shanghai Area
Shanghai is one of China's most important economic and financial centers. Since the early 1990s, along with the opening and extensive urban construction of Pudong district, Shanghai has become the locomotive of the new Chinese economy.

Pearl River Delta
The Pearl River Delta was the first area to be developed in China upon her economic reform and opening. Because of its very tight trade and financial connections with Hong Kong, the delta has become extremely important for China. Population here is highly dense, and development activity has been very extensive. The pattern of land-cover change of this area has its own specific characteristics.

Results and Discussion

Temporal-Spatial Tendencies of NDVI Variations in China

China Nationwide
A countrywide PCA was undertaken in order to detect the overall tendencies of temporal-spatial variations of NDVI in China. Using 216 monthly Maximum value composite NDVI images from 1982 through 1999, an overall spatial pattern has been identified through component 1, which is coincident with the vegetation distribution of China (Hou, 1982). That is, in the eastern part of China, NDVI presents higher values, particularly in the southwest and northeast where natural vegetation is dense and comparatively less influenced by human activity. In the west, NDVI is much lower, where infertile deserts and high elevation plateaus are dominant. Further, the temporal curve of component 2 shows a periodic change of NDVI, which is typically influenced by the change of seasons, with a peak in July and August and a trough in December and January within each year. Based on the China Vegetation Distribution Map (Hou, 1982), it is clear that component 2 is showing natural deciduous vegetation in China. That is, in north China, especially in the northeastern area, the seasonal change of NDVI strongly corresponds to the one-peaked seasonal fluctuation characterized by the temporal curve of component 2, yet southward, the trend breaks gradually. On the other hand, there are regions uncorrelated with this seasonal fluctuation, which reveals the regional diversity of NDVI change patterns in China. As a result, different regions command different factor composites that conduct differential temporal-spatial distributions of NDVI. Component 3 shows a different temporal and spatial pattern. It has been identified to indicate a cropping pattern, based on the temporal and spatial analysis. This is an area of wheat and rice agriculture (Hou, 1982). Therefore, on the countrywide scale, PCA is able to show the overall photosynthesis activity in China as well as showing major phenological trends involving both natural deciduous vegetation and agriculture vegetation.

Sub-regions
As is well known, component 1 generally indicates overall NDVI distribution. Therefore, the illumination below is chiefly focused on later components.

Northeast China
The temporal curve of component 2 shows a highly regular fluctuation in each year, with a peak in December and January and a trough in July and August. The spatial dimension of component 2 shows that nearly the entire region is strongly negatively correlated to the tendency of the temporal curve, which reveals that this region's NDVI variation indeed follows such a uni-peaked progression, but in the reverse direction as shown in the curve. That is, the maximum NDVI value occurs in the period of July to August instead of December to January. This pattern clearly corresponds to the pattern addressed by the temporal curve of component 2 for the entire China above. Compared to the China Vegetation Distribution Map (Hou, 1982), the pattern can be explained by the characteristics of climate, natural vegetation, and agricultural cropping structure of this region. In this region the winter is frigid, but summer's average temperature is generally over 24°C. Its adjacency to the Pacific Ocean makes the yearly average precipitation reach 700-800 mm (Liu, 1985). The natural vegetation type is forest, composed of masses of conifers and broad-leaf deciduous trees.

The Daxing'an and Changbei-shan Mountains in this area are in China's largest natural forest zone. It can be concluded that it is the specific characteristics of climate and vegetation that make this region's NDVI obviously follow the uni-peaked change pattern within one year. In summer, NDVI goes up to its maximum value along with the luxuriant growth of vegetation and increase of rainfall; in winter, NDVI declines to its harsh winter environment. On the other hand, this region's agricultural cropping structure in large part contributes to the explanation of the uni-peaked pattern as well. Restricted by temperature as well as precipitation conditions, crops in this area can only be harvested once per year. Crops are seeded in late spring, and harvested in early autumn (Liu, 1985). In late July to August, crops enter the most luxuriant growth period, which directly leads to the maximum NDVI value. It is the temporal-spatial coincidence of variations of natural vegetation and crops along with seasonal change that makes the variation of NDVI of this area concerted with the repeated uni-peaked fluctuation rhythm over years.

Yellow River Valley
The temporal curve of principal component 2 of this area exhibits a two-year period in which there are two big peaks and one small peak. The two big peaks occur in August to September, separately in each year, and the small peak grows from November of the previous year and reaches its crest in late February to March of the following year. Spatially, Weihe basin, Henan province, Anhui province, and north Jiangxu province, all located in the middle-lower reach of the Yellow River, show significant positive correlation with the temporal curve. On the other hand Shanxi , Inner Mongolia, Ninxia, and Shanxi provinces at the upstream exhibit a negative correlation. By reference to China Vegetation Distribution Map (Hou, 1982), it is found that the former is subject to the agricultural zone with three harvests every two years. Summer and fall months ranging from May to November are the growth periods of wheat and rice, and the stage spanning November of the previous year to May of the following year is that of winter wheat (Liu, 1985). The upstream area is principally covered by semi-arid grasslands. It can be concluded that the PCA methodology can clearly pull out agricultural zones under different phenological conditions.

Yangtze River Valley
From the PCA results of Yangtze River Valley, it is concluded that the NDVI change pattern in Yangtze River Valley is significantly different from that in Northeast China and that in the Yellow River Valley. In Yangtze River Valley, both natural vegetation and corps are broadly distributed within the area. Roughly, croplands are principally found on the north side of the Yangtze River, whereas natural sub-tropical evergreen broad-leaved forests are dominant on the south side. By reference to the China Vegetation Distribution Map (Hou, 1982), the spatial dimension of component 2 characterizes the geographic distribution of natural vegetation of this area. As opposed to component 2, component 3 spatially presents almost a reverse pattern. The temporal curve of the component 3 exhibits typically two peaks per year, one of which occurs in March to April, and another in August to September. Spatially, north Hunan province around Lake Dongting, Hubei province, Anhui province, Jiangsu province, and Zhejiang province are significantly correlated to the curve. The southern part, however, generally exhibits negative or non- correlation. Referring to the vegetation distribution map (Hou, 1982), the spatial pattern is very highly coincident with the spatial distribution of rice, which, seeded in late January to February and late June to July, respectively, and harvested in May to early June and November to early December correspondingly, is the overwhelming crop type in this area (Huang, 1994 and Liu, 1985). For this, it can be concluded that component 3 represents the part of NDVI change caused by the growth of rice. In addition, the reasons that the southern part of the region presents negative or non-correlation stem from overwhelming sub-tropical natural vegetation there as well as sparse crops with three harvests in one year (Hou, 1982).

Identification of Natural and Human Factors
The temporal-spatial change pattern of NDVI is not only controlled by large-scale factors, e.g., seasonal shifts and regional agricultural cropping structures as addressed above, but also significantly influenced by some small-scale or short-term factors, especially in smaller regions or during some specific periods. In order to identify how these small-scale natural and artificial factors influence the change of NDVI, the author further applied PCA to Lake Dongting, Shanghai, and Pearl River Delta (Guangzhou Area) separately.

Flood Hazards in Lake Dongting
Floods in this area generally take place in May to July. The author therefore extracted the data series of this stage for PCA. However, the resulting components did not reveal significant NDVI change due to the influence of floods. Again, the author attempted to analyze the data series of August and September separately. As a result, the flood accidents can be clearly detected in the first component of the series in both August and September. The resulting temporal curve of the August series shows that the NDVI value of 1997 was much lower than that of any other year, and the same was true of the September series. Why could the series of May to July not identify the flood accidents, but the series for either August or September could? This involves regional climate conditions and the growth pattern of crops. During the period of late May to July, summer crops enter their harvest stage. The croplands are either uncultivated temporarily or just sparsely covered by new rice seedlings. The result is that the NDVI value of this stage is normally much lower. Hence it is difficult to distinguish the NDVI change of this stage because the NDVI of water approximates that of bare lands. Furthermore, cloud cover easily influences the imagery of this stage as well. In August and September, fall crops are normally entering their most luxuriant growth season, and their NDVI should normally approach its maximum. When floods occur, crops will generally be subject to fatal devastation. As a result, the NDVI of this stage in a flood year is much lower than its normal value. Therefore, it is concluded that August and September are the most appropriate months in which to check the NDVI change due to the influence of flooding. Further, the author attempted to delineate the possible flooded extent according to pixels' spatial correlations with the flood accident curve. The method used is to reclassify the spatial characteristic map of component 1 of the September series. Thus, possible flooded areas during 1995 through 1999 may be distinguished. As shown in Map 14, pixels in red represent the highly possibly flooded area, which are at the top 50% of positive index, showing strongly significant correlations to the curve. Pixels in yellow characterize the slightly possibly flooded area, which fall into the lower 50% of positive index, showing weak correlations to the curve. Pixels in green address the impossibly flooded area, which present no or negative correlations to the curve.

Urbanization in Shanghai Area
Shanghai's current development pattern follows a west-to-east mode. Divided by Huangpu River, the western part of the city is called Puxi, and the eastern Pudong. Puxi has been the old center of the city since 1843. Though Puxi's construction has not ever been suspended since the early 1990s, the construction focus has shifted to the Pudong district. As a result, Shanghai is once again becoming the biggest economic and financial center of China, and creating a so-called "big economic triangle" with Hong Kong and Taiwan (Zhou, 1997).

As opposed to the extensive construction in urban areas, Chongming Island, the third biggest island in China, located at the estuary of Yangtze River, is becoming the most important agricultural base of Shanghai area. The yield of Chongming Island aims at agricultural product markets. Constructing areas solely for agricultural use is its target. Meanwhile, the local government has designed and constructed a 360-hectare man-made forest garden. These policies not only maintain the original natural landscape of Chongming Island, but also make it a new resort for vacation and entertainment.

The PCA for Shanghai area is intended to detect the impacts of human development either resulting in deforestation or reforestation upon NDVI variations. Through the PCA of the series of 1982-1999 yearly average NDVI value, it is found that the continuously rising temporal curve of component 3 can well characterize the progression of urbanization. By comparison and contrast of its spatial characteristic map with its temporal curve, it is clear that urbanization is expanding around the old center of the city in a radial pattern toward the outlying areas. The central zone spatially is significantly negatively correlated to the temporal curve, whereas the spatially negative correlations of the adjacent zones with the curve are gradually dropping down. This tendency also implicitly shows a more rapid urbanization in Pudong than in Puxi, considering that Pudong was originally a large piece of undeveloped cropland in the 1980s. This urban expanding pattern is highly correspondent to the present-day Shanghai's development mode. It is apparent that urbanization is encroaching upon a large amount of croplands and natural forest cover, resulting in the dramatic decrease of NDVI value. As opposed to the development in the central urban area, Chongming Island is experiencing an utterly reverse progression. Spatially, this area is significantly correlated to the rising temporal curve, which is in keeping with its agricultural development mode and reforestation commitment (Chang, 1999).

Urbanization in the Pearl River Delta
Since the early 1980s, along with the rapid opening of China's economy, the Pearl River Delta has become one of China's most vigorous economic zones. Similar to that of Shanghai, the urbanization tendency of the Pearl River Delta can be represented by component 3. Nevertheless, the differences of these two areas are obvious, which is shown explicitly both in the temporal progress and in the spatial expanding pattern of urbanization. Over time, as shown in the temporal curve of component 3, the urbanization of the Pearl River Delta experienced a very rapid jump from 1984 to 1987. This jump corresponds to the fact that Guangzhou and Shenzhen cites were rapidly developing in this period because of the rapid growth of trade with Hong Kong. Since then, the urbanization has continued but gradually been slowing down. This tendency is partly due to the shift of China's economic focus to the Shanghai area (Yang, 1994). Spatially, as opposed to Shanghai's pattern, the urbanization in the Pearl River Delta follows a multiple-center expanding model. That is, urban areas stretch outwards separately around Guangzhou, Shenzhen, Foshan, and Huizhou and their satellite cities, such as Huadu, Chonghua, and Jiangmen. These cities typically comprise a complete, huge city collection in which cities join together seamlessly.

Conclusions
This thesis investigates the temporal-spatial change pattern of NDVI in China using PCA/TSA. The results have proved that PCA/TSA is a very effective method in which to identify both macro and micro factors driving the change of NDVI. In particular, the thesis paves a way to detect the impacts of extreme physical accidents and human-induced activities upon the NDVI change. The analytical results are quite exciting and satisfactory.

Bibliography

• Agbu, P. A., and M.E. James (1994). The NOAA/NASA Pathfinder AVHRR Land Data Set User's Manual. Greenbelt, Maryland, Goddard Distributed Active Center Publication.
• Anyamba, A., Eastman, J.R, and Fulk, M. (1995). NDVI Image Time Series as an Indicator of Climate Variability. Worcester, Clark University.
• Anyamba, A., and Eastman, J.R. (1996). "Interannual Variability of NDVI over Africa and its Relation to El Nino / Southern Oscillation." International Journal of Remote Sensing (17): 2533-2548.
• Box, E. O., and Holben, B., and Kalb, V. (1989). "Accuracy of The AVHRR Vegetation Index as a Predictor of Biomass, Primary Productivity and Net CO2 Flux." Vegetatio (80): 71-89.
• Chang, S. C. (1999). Investment and Development of Chongming County.Chen, H., and Qiu, J.W. (1994). China Urbanization (Chinese). Bejjin.Eastman, J. R., and McKendry, J.E., and Fulk, M. (1995). United Nations Institute for Training and Research. Worcester, Clark Labs.
• Eastman, J. R. (1999). Guide to GIS and Image Processing. Worcester, Clark Labs.
• Gates, D. M., and Keegan, H.J., and Schleter, J.C., and Weidner, V.H. (1965). "Spectral Properties of Plant." Applied Optics (4): 11-20.
• Goward, S. N., and Tucker, C.J., and Dye, D.G. (1985). "North American Vegetation Patterns Observed with NOAA-7 Advanced Very High Resolution Radiometer." Vegetatio (64): 3-14.
• Hastings, D. A., and Emery, W.J. (1992). "The Advanced Very High Resolution Radiometer (AVHRR): A Brief Reference Guide." Photogrammetric Engineering and Remote Sensing (58): 1183-1188.
• He, B. C. (1991). China on the Edge, San Francisco: China Books and Periodicals, Inc.
• Hou, X. (1982). China Vegetation Distribution Map.
• Huang, P. C. C. (1994). The Peasant Family and Rural Development in Yangtze Delta, 1350-1988 (Chinese), Oxford University Press.
• Li, Jingbao, H. Z., Duan, Zhengliang (1997). "Hazardous Floods and Flood Disaster of Dongting Lake Area. Economy Geography (Chinese)." 169-170.
• Lillesand, T. M., and Kiefer, R.W. (1994). Remote Sensing and Image Interpretation. New York, John Wiley & Sons.
• Liu, H. (1985). China Geography -The Third Volume (Chinese), National Translator-Editor Embassy.
• Xiang, L. (1999). "Reflection on the Hunan and Hubei '98 Flood Survey (Chinese)." Journal of China Institute of Water Resource and Hydropower Research 3(2): 80-85.
• Sellers, P. J. (1985). "Canopy Reflectance Photosynthesis and Transpiration." International Journal of Remote Sensing (6): 1335-1372.
• Cai, Shuming C. Z. (1992). "Strategy Against Floods and Water-logging of Middle Yangtze River (Chinese)." Chinese Academy Sciences, Geoscience Board, China Natural Disaster and Mitigation: 379-380.
• Tucker, C. J., and Townshend, J,R,G,, and Goff, T.E. (1985). "African Land-cover classification using satelite data." Science (227): 369-375.
• Wang Kelin, Z. C. (1999). The Mechanism of Flood and Water-logging Disaster Formation, Ecological Countermeasures of Reducing Disasters and Basin Management. Changsha, Changsha Institute of Agricultural Modernization: 1-13.
• Yang, Gan Z. Q. (1994). The political economics of China reform (Chinese), Oxford University Press.
• Young, S. S. (1997). The Use of Coarse-Grain, Global-Scale Remote Sensing Data to Analyze Forest Cover, Extent and Change, in China, 1982 - 1992. Department of Geography. Worcester, Clark University.
• Young, S. S. (1998). Time Series Land-Cover Change Analysis Using Principal Components Analysis and AVHRR NDVI Data. Salem, Salem State College.
• Young, S. S. and Wang, C.Y. (2000). "Land-cover change analysis of China using global-scale Pathfinder AVHRR Landcover (PAL) data, 1982-92." International Journal of Remote Sensing 00(00): 0000-0000.
• Zhou, Q. (1997). Difference Investigation of China Area Development (Chinese), Oxford University Press.